| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Departments of Cell Biology and
b Obstetrics & Gynecology, University of Virginia Health System, Charlottesville, Virginia 22908
c Ludwig Institute for Cancer Research, University College Branch, London, United Kingdom W1P 8BT 91
ABSTRACT
Cancer-testis antigens (CTAs) represent potential targets for cancer immunotherapy because these proteins are widely distributed in tumors but not in normal tissues, except testes. In this paper, we identify homology of the CTA CTp11 with SPAN-X (sperm protein associated with the nucleus mapped to the X chromosome). On two-dimensional Western blots of human sperm extracts, SPAN-X antibodies recognized 19 spots ranging from 20 to 23 kDa with isoelectric points from 5.0 to 5.5. Differential extraction of spermatozoa demonstrated that the SPAN-X protein is highly insoluble. Only 50% of ejaculated spermatozoa exhibited SPAN-X immunofluorescent staining. Dual localization of the sex chromosomes and the SPAN-X protein demonstrated that an equal number of X- and Y-bearing spermatozoa exhibited SPAN-X staining. In transfected mammalian CV1 cells, the SPAN-Xa and SPAN-Xb proteins were localized to the nucleus and cytoplasm, respectively, by indirect immunofluorescence. On immunoblots of CV1 cells, the SPAN-Xa protein migrated at 15–20 kDa, whereas the SPAN-Xb protein migrated at a higher molecular weight of 21–22 kDa. The SPAN-X protein was ultrastructurally associated with nuclear vacuoles and the redundant nuclear envelope. SPAN-X is the first protein specifically localized to these poorly characterized structures of the mammalian sperm nucleus and provides a unique biochemical marker for investigation of their function in spermatozoa as well as the role of SPAN-X/CTp11 in human tumors.
sperm, sperm maturation, spermatogenesis, testis
INTRODUCTION
Cancer/testis antigens (CTAs) are defined as immunogenic antigens that are expressed in the normal testis and a variable proportion of human cancers [1–3]. Common features of these antigens include 1) mRNA expression in normal testis but typically not in other noncancerous tissues; 2) mRNA expression in a wide range of human tumor types; 3) expression in malignancies in a lineage-independent manner; 4) existence of multigene families; and 5) mapping of the gene, with one exception, to the X chromosome. Although the functions of other tumor differentiation antigens are known, little is known about the localization and function of CTAs in either normal testis or tumor cells.
From an immunological perspective, CTAs represent targets for immunotherapy because these proteins are widely distributed in a number of tumors, but among normal tissues are mainly present only in the testis. Due to the blood-testis barrier that normally limits contact between testicular and immune cells and to the lack of human leukocyte antigen (HLA) class I expression on the surface of germ cells, the testis is considered an immune-privileged site [4]. These features suggest that only cancerous cells will be targeted by cytotoxic T lymphocytes (CTLs) during CTA immunotherapy; thus, these proteins are lead candidates as vaccines for the treatment of tumors.
Recently, a new CTA, designated CTp11 (cancer/testis-associated protein of 11 kDa), was identified by comparing mRNA expression between a human parental melanoma cell line and its metastatic variant using mRNA differential display [5]. Northern blot and reverse transcription-polymerase chain reaction (RT-PCR) analyses demonstrated that CTp11 mRNA was expressed in highly metastatic cell lines of melanomas and bladder carcinomas as well as tumors of lung, breast, colon, bladder, testis, and melanoma, but not in normal tissues except testis. The CTp11-green fluorescent protein fusion protein had a deduced mass of 11 kDa and was localized to the nucleus of transfected COS cells. The CTp11 gene was mapped to the X chromosome by homology with genomic clones localized to the X chromosome and by genomic PCR analysis of a panel of rodent/human hybrid cell lines containing specific human chromosomes. The X chromosomal gene localization and the expression of CTp11 in normal testis, tumor cell lines, and tumors establishes CTp11 as a CTA [5].
In a previous report, we described two related cDNA and peptide sequences of SPAN-X, a sperm protein associated with the nucleus mapped to the X chromosome, the SPANX genomic structure, the pattern of mRNA expression, and the localization of SPAN-X protein to the human sperm head [6]. The two related SPAN-X peptide sequences, designated SPAN-Xa and SPAN-Xb, contained three overlapping consensus nuclear localization signals, a high percentage (33%–37%) of charged amino acid residues, and a relatively acidic pI (isoelectric point) (4.88–6.05). Northern analysis of mRNA from 50 human tissues identified a 0.6-kilobase (kb) SPAN-X transcript exclusively in the testis and in situ hybridization of human testis sections showed SPAN-X mRNA expression in haploid round and elongating spermatids. The SPANX gene was mapped to chromosome Xq27.1 by fluorescence in situ hybridization (FISH) and by Southern analysis of human/mouse somatic cell hybrids. On Western blots of human sperm proteins, anti-recombinant SPAN-X antibodies reacted with broad bands migrating between 15–20 kDa. Immunofluorescent labeling demonstrated SPAN-X localization to nuclear craters and cytoplasmic droplets of ejaculated human spermatozoa.
Homologies between the SPAN-Xa, SPAN-Xb, and CTp11 sequences are identified in this report. To further characterize the SPAN-X/CTp11 CTA protein in normal cells, antisera raised against recombinant SPAN-X were employed to examine the protein's microheterogeneity, define its solubility characteristics, and localize SPAN-X at the fine structural level in human spermatozoa. Using immunofluorescence microscopy, we examined the distribution of SPAN-Xa and SPAN-Xb proteins in transfected CV1 cells and demonstrated that the two SPAN-X variants translocate to different subcellular compartments. It is interesting that while only 50% of ejaculated human spermatozoa exhibited SPAN-X immunostaining, the SPAN-X protein was equally distributed between X- and Y-bearing spermatozoa. To our knowledge, SPAN-X is the first protein reported to segregate to a subpopulation of spermatozoa. These biochemical and cellular characteristics of the SPAN-X/CTp11 protein in its normal site of expression, the spermatid and spermatozoon, may shed additional light on its role in highly metastatic human tumors.
MATERIALS AND METHODS
Antibodies
To produce polyclonal antisera (pAb), mice and guinea pigs were immunized with recombinant SPAN-X (re-SPAN-X) as described in a previous study [6]. All animal investigations were conducted in accordance with the Guide for Care and Use of Laboratory Animals.
Sperm Preparation and Extraction
Semen specimens were donated by normal, healthy men. Only ejaculates with normal semen parameters [7] were used in this study. All samples were obtained with informed consent using forms approved by the University of Virginia Human Investigation Committee. Individual semen samples were allowed to liquefy at room temperature and were washed by centrifugation. For some experiments, mature spermatozoa were separated from seminal plasma, immature germ cells, and nonsperm contaminating cells (mainly white blood cells and epithelial cells) by the swim-up technique [8]. In other experiments, mature spermatozoa were separated by Percoll (Pharmacia Biotech, Piscataway, NJ) density gradient centrifugation [9]. For two-dimensional electrophoresis, human sperm pellets were prepared as described [10].
To assess relative protein solubility, washed spermatozoa were extracted in 0.5% 3-[(3-cholamidopropyl)dimethlammonio]-1-propane-sulfonate (CHAPS) in Tris-buffered saline (TBS; 10 mM Tris-HCl pH 7.5, 150 mM NaCl, and protease inhibitor cocktail [Roche Molecular Biochemicals, Indianapolis, IN]) with or without 600 mM KCl, for 2 h at 4°C. In some experiments, 2 mM dithiothreitol (DTT) was added to the extraction buffer. The suspension was then centrifuged at 10 000 x g for 20 min. Both the supernatant fluid and pellet fractions were utilized for SDS-PAGE and Western blotting as described below.
Protein concentrations were determined using the Pierce BCA method (Pierce Chemical Co., Rockford, IL) according to the manufacturer's specifications employing BSA as a standard, by the spectrophotometric method, or both [11].
Polyacrylamide Gel Electrophoresis and Western Blotting
For one-dimensional SDS-PAGE, electrophoresis was performed on 15% acrylamide gels [12] with 50 µg of protein per lane. Following SDS-PAGE, polypeptides were transferred onto a nitrocellulose membrane [13]. For two-dimensional electrophoresis, isoelectric focusing (IEF), SDS-PAGE, and electrotransfer were performed as previously described [10].
Western blots were incubated in PBS containing 0.05% Tween-20 and 5% nonfat dry milk to block nonspecific protein-binding sites. Blots were washed with PBS containing 0.05% Tween-20 between all subsequent incubation steps. Blots were incubated in preimmune or postimmune sera diluted in PBS-normal donkey serum (PBS-NDS) followed by horseradish peroxidase (HRP)-conjugated F(ab')2 fragments of donkey anti-guinea pig immunoglobulin G (IgG). HRP conjugates were visualized using TMB reagent following the manufacturer's protocol (Kirkegaard & Perry Laboratories, Gaithersburg, MD).
Immunofluorescence Microscopy
Swim-up spermatozoa were fixed at 4°C with 2% formaldehyde (Electron Microscopy Sciences, Ft. Washington, PA) in PBS for 20 min. In some experiments, the cells were not fixed with formaldehyde. Fixed or nonfixed cells were air-dried onto slides and washed three times with PBS. To permeabilize the cells, the slides were incubated in methanol and washed with PBS. Nonspecific protein binding sites were blocked by incubating the slides in PBS with 10% NDS or normal goat serum (NGS). Slides were incubated with either preimmune mouse or guinea pig serum (1:250), immune mouse serum or guinea pig serum (1:250) diluted in PBS with 1% NGS or NDS (PBS-NS). The slides were washed and incubated with FITC-conjugated F(ab')2 fragments of goat anti-mouse IgG/IgM (1:200) or donkey anti-guinea pig IgG (1:200; Jackson ImmunoResearch, West Grove, PA) in PBS-NS. For confocal microscopy, sperm DNA was stained with propidium iodide during incubation with the secondary antibody. Slides were washed with PBS and mounted with SlowFade Light (Molecular Probes, Eugene, OR). Cells were observed by differential interference contrast (DIC) and epifluorescence microscopy using a Zeiss axiophot microscope. For confocal microscopy, cells were observed using a Zeiss axiovert microscope. Digital images were obtained using Zeiss LSM software (Carl Zeiss Inc., Thornwood, NY). The proportion of cells containing positive immunostaining was calculated and the 95% confidence interval of the proportion was estimated [14].
Combined FISH and Immunofluorescent Microscopy of Human Spermatozoa
Fluorescence in situ hybridization using a green fluorochrome-labeled X (alpha satellite) or Y (satellite III) chromosome probe was performed on swim-up spermatozoa according to the manufacturer's protocols (Vysis, Downers Grove, IL) with modifications. Spermatozoa were air-dried onto slides, fixed with methanol:glacial acetic acid (3:1), and dehydrated through an ethanol series. The sperm nuclei were swollen by incubation in swelling buffer 1 (0.1 M Tris-HCl pH 8.0 and 10 mM DTT) for 30 min followed by incubation in swelling buffer 2 (50 mM Tris-HCl pH 8.0, and 10 mM 3,5-diiodosalicylic acid, lithium salt; LIS [Sigma Chemical Company, St. Louis, MO]) for 1–3 h. Denaturation, hybridization, and post-hybridization stringency washes were performed according to the Vysis protocol. The CEP X SpectrumGreen (alpha satellite) was used as a probe for the X chromosome and the CEP Y SpectrumGreen (satellite III) (Vysis) was used as a probe for the Y chromosome in different experiments.
Indirect immunofluorescent labeling for SPAN-X was performed as above except that a TRITC (rhodamine)-conjugated F(ab')2 donkey anti-guinea pig IgG (Jackson ImmunoResearch) was used as the secondary antibody. Slides were mounted with SlowFade containing DAPI II counterstain (Vysis). Cells were observed by epifluorescence microscopy using a Zeiss axiophot microscope. Individual blue, green, and red fluorescent images were obtained using a digital camera (Hamamatsu Corp., Bridgewater, NJ) and compiled using Openlab software (Improvision Inc., Boston, MA). The proportions of cells containing red, green, and red/green staining were calculated and the 95% confidence interval of each proportion was estimated [14].
Immunoelectron Microscopy
For postembedding immunolabeling, washed human spermatozoa were fixed on ice with 4% formaldehyde, 0.5% glutaraldehyde in 0.1 M sodium phosphate buffer pH 7.4, rinsed in buffer, dehydrated through an ethanol series, and embedded in Lowicryl K4M resin (Electron Microscopy Sciences). Thin sections were mounted on nickel grids and immunostained. Primary antibodies were used at a concentration of 1:50 (mouse antiserum) or 1:200 (guinea pig antiserum). Five-nanometer gold-conjugated secondary antibodies (Goldmark, Phillipsburg, NJ) were employed at a concentration of 1:50. Gold-conjugated antibodies used included F(ab')2 fragments of goat anti-mouse IgG and whole IgG of goat anti–guinea pig IgG. Grids were rinsed with PBS, rinsed with water, stained with uranyl acetate, and carbon-coated.
Transfection of Mammalian Cells
To express and localize SPAN-X in mammalian cells, the SPAN-Xa and SPAN-Xb open reading frames (ORFs) were subcloned into the mammalian expression vector, pcDNA3.1, and used to transfect CV1 cells (African green monkey kidney cells). PCR primers at the 5' and 3' ends of the SPAN-X ORF were designed containing BamHI and XbaI restriction sites (italics), respectively. The 5' primer used was 5'-CGGGATCCCCTRCYRYWGACATYGAAGAACC-3' and the 3' reverse complement primer used was 5'-GCTCTAGAGCCSAAGKTTGAGRGATGT-3'. The SPAN-Xa and SPAN-Xb fragments were PCR-amplified, gel purified, digested with BamHI and XbaI, and subcloned into the pcDNA3.1 vector according to the manufacturer's protocol (Invitrogen, Carlsbad, CA). The plasmid constructs (pcDNA/SPAN-Xa and pcDNA/SPAN-Xb) were verified by automated sequencing. The pcDNA vector alone, pcDNA/SPAN-Xa, and pcDNA/SPAN-Xb constructs were transfected into CV1 cells (American Type Culture Collection, Manassas, VA) using the TransIT reagent (PanVera, Madison, WI) according to the manufacturer's protocols. Transfected cells were used for immunofluorescent labeling and Western blot analyses.
RESULTS
SPAN-X Is Homologous to the Cancer/Testis-Associated Protein, CTp11
Previously, we reported the cDNA and peptide sequences of SPAN-Xa and SPAN-Xb (GenBank accession numbers AF098306 and AF098307, respectively) [6]. BLAST analysis of the SPAN-X sequences subsequently identified a recently published homologue, designated CTp11 for cancer/testis associated protein of 11 kDa (GenBank accession number AJ238277) [5]. The CTp11 cDNA sequence was 98% and 93% identical to the SPAN-Xa and SPAN-Xb cDNA sequences, respectively. The CTp11 peptide sequence was 97% identical to the SPAN-Xa peptide with only three amino acid substitutions (Fig. 1). The three amino acid substitutions found in CTp11 were identical to the corresponding SPAN-Xb amino acid residues. The CTp11 peptide sequence was 86% identical to the SPAN-Xb peptide. Like SPAN-Xa, the CTp11 peptide sequence did not contain the six amino acid insertion nor the putative N-linked glycosylation site present in the SPAN-Xb protein. Furthermore, several of the N- and C-terminal amino acids of the SPAN-Xa and SPAN-Xb peptides differed, including six N-terminal substitutions at amino acids 2–3 and 7–10 and five C-terminal substitutions at amino acids 94, 97–98, 100, and 102 of SPAN-Xb. Three overlapping putative nuclear localization signals (NLSs) were conserved in all three peptide sequences (Fig. 1; overlap in green). The ORFs contained two overlapping NLSs of the single basic type at amino acids 37–43 (PAPKKMK) and 39–45 (PKKMKTS) of the SPAN-Xa ORF (Fig. 1, yellow box) and a consensus, bipartite, basic type NLS (SPAN-Xa amino acids 40–57; KKxKxxxxxxxxxxRxRR; blue box) consisting of two clusters of two to four basic amino acids separated by approximately 10 spacer amino acids [15].
|
Anti-SPAN-X Antibodies Recognize Polymorphic Proteins on Western Blots
To assess the relative solubility of the SPAN-X protein, washed spermatozoa were extracted in a variety of solubilization buffers (Fig. 2). The soluble supernatant fraction and insoluble pellet fraction were examined by SDS-PAGE under reducing conditions and immunoblotting with anti-SPAN-X antibodies. On Western blots, anti-SPAN-X sera reacted with broad bands migrating between 11–12 and 15–20 kDa. SPAN-X was observed in the insoluble pellet fraction of spermatozoa extracted with 0.5% CHAPS, 0.5% CHAPS containing 600 mM KCl, and 0.5% CHAPS containing 600 mM KCl and 2 mM DTT. SPAN-X was observed in the soluble fraction only following extraction with 1% SDS. The 11- to 12-kDa immunoreactive band was not observed in the SDS supernatant or pellet fractions. These data indicate that SPAN-X is a relatively insoluble sperm protein.
|
For two-dimensional gel analysis of the SPAN-X protein, Percoll-purified human spermatozoa from six donors were extracted in lysis buffer. Sperm extracts were separated in the first dimension by isoelectric focusing (IEF), separated in the second dimension by SDS-PAGE, and either silver-stained or transferred to nitrocellulose and immunostained with the anti-SPAN-X pAbs (Fig. 3). Approximately 19 immunoreactive spots were visualized at 19.8–23.5 kDa with pIs ranging from 5.0–5.5 (Fig. 3, B–D; Table 1). The net charges of the SPAN-Xa and SPAN-Xb deduced amino acid sequences were 4.88 and 6.05, respectively, values similar to the pIs of immunoreactive spots observed on two-dimensional Western blots. On corresponding silver-stained gels, these proteins appeared yellow in color (Fig. 3, A–C). These results indicate that SPAN-X is a relatively acidic, agyrophobic, polymorphic protein.
|
|
SPAN-X Protein Localizes to 50% of Human Spermatozoa
To localize the SPAN-X antigen in human spermatozoa, indirect immunofluorescence was performed using the anti-SPAN-X pAbs and control preimmune sera. Immunofluorescence localization demonstrated intense staining in the nuclear craters (Fig. 4, large arrows; Fig. 5, a and b) and/or cytoplasmic droplets (Fig. 4, small arrow; Fig. 5, c–f) of formaldehyde-fixed, methanol-permeabilized, swim-up spermatozoa with each postimmune antisera. SPAN-X staining revealed several phenotypes, including small and large nuclear craters (Fig. 5, a and b, respectively); multiple craters (Fig. 5a); small and large cytoplasmic droplets (Fig. 5, c and d, respectively) without crater staining; and both cytoplasmic droplets and nuclear craters (Fig. 5, e and f). Identical localization was observed on paraformaldehyde-fixed or nonfixed, air-dried spermatozoa, indicating that the localization is not an artifact of aldehyde fixation (data not shown). Immunofluorescent staining was not observed on every spermatozoon in the field of view although nuclear craters, cytoplasmic droplets, or both were present in these cells (Fig. 4, circled area). Spermatozoa incubated with preimmune sera exhibited no fluorescent staining (Fig. 4, right panels).
|
|
The incidence of SPAN-X immunofluorescent staining was categorized according to SPAN-X phenotypes (Table 2). Using DIC imaging, greater than 95% of spermatozoa exhibited one or more nuclear craters. Of 1281 spermatozoa examined from a pool of 11 donors using indirect immunofluorescence to localize the SPAN-X protein, 20.4% of spermatozoa exhibited one or more immunofluorescent craters without cytoplasmic droplet staining (Fig. 5, a and b), 25.7% of spermatozoa showed staining of the cytoplasmic droplet without crater staining (Fig. 5, c and d), and 4.5% showed both crater and cytoplasmic droplet staining (Fig. 5, e and f). In total, the SPAN-X protein localized to either nuclear craters, cytoplasmic droplets, or both in 50.6% (with a 95% confidence interval of ± 2.74%) of spermatozoa by indirect immunofluorescence using conventional microscopy (Table 2).
|
To further investigate the observation that the SPAN-X protein was present in 50% of the sperm population, fluorescent-labeled, swim-up spermatozoa from a pool of 15 donors were analyzed by confocal microscopy through all planes of the sperm head to examine the incidence of SPAN-X staining. Spermatozoa were immunolabeled with an FITC-conjugated secondary antibody and the sperm DNA was stained with propidium iodide. Z-sections (0.5 µm) were taken through the depth of the sperm head by extended depth of focus. Of 1039 spermatozoa examined by digital imaging, 515 (49.6% ± 3.0%) exhibited immunofluorescent staining with the anti-SPAN-X antibody (Table 2). In total, 2320 spermatozoa were examined by either conventional or confocal microscopy. Of these 2320 spermatozoa, 1163 (50.1% ± 2.0%) exhibited immunofluorescent labeling with the SPAN-X antisera (Table 2).
To examine the range of SPAN-X-containing spermatozoa in individual donors, immunofluorescent staining was scored for each of 28 donors. Greater than 100 cells were scored per donor and 3442 spermatozoa were scored in total. The range of spermatozoa exhibiting SPAN-X staining in these individual donors was 40.6%–55.2%. The mean and standard deviation of SPAN-X-positive cells in this experiment were 46.8% ± 4.1%.
To determine if SPAN-X segregated with X- or Y-bearing spermatozoa, the number of X- and Y-bearing spermatozoa that contained SPAN-X protein was studied by both FISH to identify the X or Y chromosome and by immunofluorescence to detect SPAN-X using swim-up spermatozoa with swollen nuclei (Fig. 6A). The number of cells exhibiting SPAN-X and X chromosome staining, SPAN-X without X staining, no SPAN-X with X staining, and neither SPAN-X nor X staining were counted, the proportions calculated, and the 95% confidence interval determined for each subset (Fig. 6B). In total, 389 cells were counted and scored. Of these, 24.9% exhibited SPAN-X and X chromosome staining while 27.2% exhibited SPAN-X without X staining. These data indicated that SPAN-X protein was equally distributed between X- and Y-bearing spermatozoa. To confirm this observation, the numbers of cells exhibiting SPAN-X and Y chromosome staining, SPAN-X without Y staining, no SPAN-X with Y staining, and neither SPAN-X nor Y staining were also counted, the proportions calculated, and the 95% confidence interval determined for each subset (Fig. 6C). In total, 343 cells were counted and scored. Of these, 24.8% exhibited SPAN-X and Y chromosome staining, whereas 22.4% exhibited SPAN-X without Y staining.
|
SPAN-X Protein Localizes to Nuclear Vacuoles and Redundant Nuclear Envelope
Spermatozoa prepared for postembedding immunoelectron microscopy exhibited a number of randomly distributed cavities throughout the condensed chromatin of the sperm head (Figs. 7 and 8). These cavities, identified as nuclear vacuoles, are non-membrane-bound areas filled with amorphous, granular material and are devoid of nuclear chromatin. Spermatozoa immunolabeled with the anti-SPAN-X antibodies showed gold particles over granular material within nuclear vacuoles (Fig. 7B; Fig. 8, A through C), although not all vacuoles were labeled. Gold labeling was often associated with membranous material within the nuclear vacuoles (Fig. 8C). Few particles were observed in the surrounding nucleoplasm. Often, SPAN-X immunostaining was observed in nuclear vacuoles that appeared to open into the subacrosomal or perinuclear spaces (Fig. 8B). In addition, specific staining of folds in the extensive redundant nuclear envelope was observed in the cytoplasmic droplet at the base of the sperm head (Fig. 7, B and C) in some spermatozoa. Staining on the redundant nuclear envelope was observed caudal to the boundary of the posterior ring. Spermatozoa labeled with preimmune sera showed no staining of the nuclear vacuoles, redundant nuclear envelope, or other sperm organelles.
|
|
SPAN-Xa Localizes to the Nucleus in Transfected Mammalian CV1 Cells
To express and localize SPAN-X in mammalian cells, the SPAN-Xa and SPAN-Xb ORFs were subcloned into the mammalian expression vector, pcDNA3.1, and used to transfect CV1 cells (African green monkey kidney cells). The expressed protein was detected by indirect immunofluorescent labeling of transfected cells with anti-SPAN-X antibodies. The SPAN-Xa protein localized to the nucleus of transfected CV1 cells (Fig. 9A). In contrast, the SPAN-Xb protein was randomly distributed within the cytoplasm of the cells (Fig. 9B). CV1 cells transfected with the pcDNA3.1 vector alone exhibited no fluorescent staining (not shown).
|
To examine the protein expressed by the transfected CV1 cells, cell extracts were prepared, subjected to SDS-PAGE, and immunoblotted with anti-SPAN-X antibodies (Fig. 10). On Western blots of sperm proteins used as a positive control, postimmune sera strongly reacted with broad bands migrating between 15 and 20 kDa. Extracts of CV1 cells transfected with the SPAN-Xa expression construct showed immunoreactive bands of similar molecular mass to those observed in the sperm extract. Extracts of CV1 cells transfected with the SPAN-Xb expression construct showed immunoreactive bands of approximately 21–22 kDa. CV1 cell extracts containing vector only showed no immunoreactive proteins.
|
DISCUSSION
SPAN-X Is a Highly Insoluble, Acidic, Polymorphic Sperm Protein
Analysis of SPAN-X sequences identified a recently published homologue, designated CTp11 for cancer/testis-associated protein of 11 kDa [5]. CTp11 mRNA expression was seen in 70%–75% of melanoma tumors but not in normal tissues, except testis. Expression of SPAN-X/CTp11 in tumors and normal testis establishes this protein as a new member of the group of CTAs [5]. From an immunological perspective, CTAs represent potential targets for immunotherapy because they are widely distributed in a number of tumors but not in normal tissues, except the immune-privileged testis [1–3]. To this end, characterization of this newly identified CTA, SPAN-X/CTp11, is of considerable importance.
This study demonstrated several molecular and biochemical properties of the CTA protein, SPAN-X/CTp11, an acidic, structural protein of the human sperm nucleus. Differential extraction of human spermatozoa demonstrated that the SPAN-X protein is insoluble in zwitterionic detergents, high salt, and reducing agents but can be solubilized by denaturing detergents. The insolubility of SPAN-X was not due to inaccessibility of the sperm nucleus to extraction buffers, as the majority of SPAN-X is associated with the redundant nuclear envelope of the cytoplasmic droplet. The ultrastructural localization of the insoluble SPAN-X protein suggests that SPAN-X is a structural component of the sperm nuclear envelope or is associated with structural components of the nucleus, possibly the nuclear matrix. On Western blots of sperm proteins extracted in nonionic buffers, anti-SPAN-X sera reacted with broad bands migrating between 11–12 and 15–20 kDa. This lower molecular weight (11–12 kDa) protein was not observed on Western blots of SDS-extracted spermatozoa. Although we cannot completely rule out proteolysis, it seems unlikely because a protease inhibitor cocktail was included in all extraction buffers. The 11- to 12-kDa protein may represent SPAN-X protein that has not been post-translationally modified because it is similar in molecular weight to that of the amino acid sequences from the SPAN-X cDNAs.
Nineteen variants of SPAN-X were detected on two-dimensional immunoblots over a pI range of 5.0–5.5 and a molecular mass of approximately 20–23 kDa. The differences in apparent molecular mass observed by one- and two-dimensional SDS-PAGE are consistent with differences in sperm extraction and electrophoretic conditions. The observed pI range is consistent with the theoretical pI of the SPAN-Xa- and SPAN-Xb-deduced amino acid sequences that contain a large number of negatively charged amino acids (19%). The variations in pI and molecular mass of SPAN-X may be the result of proteolysis, post-translational modifications such as phosphorylation or glycosylation, alternative splicing, or a combination of these. The polymorphism observed in the SPANX gene described in our previous study [6] represents one possible cause of SPAN-X protein microheterogeneity.
It is interesting that SPAN-X stained yellow using standard silver staining techniques. This could potentially be due to the highly charged nature of SPAN-X, although previous literature to support this hypothesis was not found. Rather, previous literature suggested that negatively charged, sialylated glycoproteins may stain yellow by this method [16]. Glycosylation of nuclear proteins has been described in somatic cells [17, 18]. Although there are several potential glycosylation sites in the deduced amino acid sequence of SPAN-X, the presence of carbohydrate moieties on the SPAN-X protein remains to be determined.
From the standpoint of considering SPAN-X as a possible marker of malignancy, the insolubility of SPAN-X in normal spermatozoa predicts that SPAN-X may not be useful as a serum marker but may be very valuable as a cytological marker. However, the solubility of the various SPAN-X isoforms in a wide range of human tumors and patient immune responses to SPAN-X remain to be explored.
SPAN-X Localizes to Sperm Nuclear Vacuoles and Redundant Nuclear Envelope
Anti-SPAN-X antibodies recognized sperm craters and cytoplasmic droplets in ejaculated spermatozoa by indirect immunofluorescence. Sperm craters are believed to correspond to either indentations on the nuclear surface, to vacuoles within the condensed chromatin of the nucleus, or both [19]. With DIC optics, the determination of craters as nuclear surface or subsurface features is not possible. Using immunoelectron microscopy, SPAN-X was identified within the sperm nuclear vacuoles. This result directly identifies nuclear craters observed by DIC imaging as nuclear vacuoles in mammalian spermatozoa. Furthermore, SPAN-X is the first example of a specific protein localized to nuclear vacuoles or redundant nuclear membrane in spermatozoa of any species.
Based on chemical composition and morphological structures, nuclear vacuoles are believed to be derived from the nucleolus of spermatocytes and spermatids [20–22]. The nucleolus, which synthesizes the major components of ribosomes, typically contains two forms of morphologically distinguishable RNA: granules, representing maturing ribonucleoprotein (RNP) particles; and fibrils, possibly a precursor to the granules. Within the nuclear vacuoles of spermatozoa, fibrils and dense fibrillar structures are often associated with the peripheral condensed chromatin. Cytochemical staining has indicated the presence of RNPs and deoxyribonucleoproteins (DNPs) in the fibrillar structures of nuclear vacuoles [22, 23].
RNPs and DNPs have also been identified in stacks of annulate lamellae in human spermatids [23]. These lamellae are closely associated with the nuclear envelope during spermatogenesis and with the redundant nuclear envelope during late spermiogenesis [24–26]. Furthermore, these membrane stacks are often found within the nucleus as the spermatid chromatin condenses [25]. The annulate lamellae may represent a specialized endoplasmic reticulum and function in nuclear-cytoplasmic transport. The SPAN-X protein localizes to the redundant nuclear envelope and to membrane structures within nuclear vacuoles of mature spermatozoa. Whether SPAN-X is a component of the annulate lamellae, the nuclear envelope of differentiating spermatids, or both, remains to be determined. The possibility that SPAN-X represents a binding protein for either nucleic acids or basic nuclear proteins should be further examined based on its localization and the highly charged nature of the SPAN-X protein.
During spermatid elongation, the cytoplasmic droplet and redundant nuclear envelope are formed as a result of reduction in cytoplasmic and nuclear volume. Furthermore, the nuclear pore complexes are relocated to the redundant nuclear envelope during spermiogenesis [27]. Exclusion of SPAN-X from the nucleus into the residual cytoplasm/cytoplasmic droplet during spermiogenesis or epididymal sperm maturation may explain the absence of SPAN-X staining in some nuclei. The co-localization of SPAN-X to both nuclear vacuoles and the redundant nuclear envelope suggests that these structures are functionally related and that in some spermatozoa or during some stage of sperm development, these two compartments are contiguous. The presence of SPAN-X protein within nuclear vacuoles that appear to open into the subacrosomal or perinuclear space is indicative of a connection between the vacuoles and the cytoplasm or the nuclear cisternae. During spermiogenesis, continuity between the nuclear vacuoles and the redundant nuclear envelope may result in transport of materials targeted for degradation such as mRNAs, proteins, or both, into the cytoplasmic droplet for disposal. Serial sectioning of human spermatids and mature spermatozoa is required to explore these relationships.
As a marker for nuclear vacuoles and redundant nuclear envelope, SPAN-X may aid in understanding the function of these structures during spermatogenesis, epididymal maturation, and transport through the female reproductive tract. The subcellular localization of SPAN-X in human spermatozoa will serve as a basis for comparative studies in human tumors. Furthermore, following fertilization, the potential role of SPAN-X in nuclear decondensation and in formation of the male pronucleus should be examined.
Distribution of SPAN-X Protein in 50% of X- and Y-Bearing Human Spermatozoa
A remarkable observation was that close to 50% of ejaculated human spermatozoa in a donor pool exhibited immunofluorescent labeling with the SPAN-X antisera. Incomplete permeabilization of the nucleus and antibody inaccessibility caused by positional effects of the spermatozoa on slides cannot be excluded as an explanation for the low percentage of SPAN-X immunoreactive craters in the first set of experiments. However, crater staining throughout the sperm head was then investigated by Z-sectioning using confocal microscopy and similar results were obtained. To our knowledge, SPAN-X is the only protein that shows a 50% segregation in ejaculated spermatozoa of any species. The range of spermatozoa containing SPAN-X among 28 individual donors varied between 40.6% and 55.2% with a mean of 46.8% ± 4.1%. This result indicates that the 50% segregation of the SPAN-X protein is likely a general phenomenon applicable to most, if not all, individual populations of human spermatozoa.
The localization of SPAN-X to 50% of spermatozoa and its X-linked expression by haploid spermatids initially suggested that SPAN-X may be associated with only X-bearing spermatozoa. However, dual labeling of spermatozoa utiliz-ing FISH for the X or Y chromosome and indirect immunofluorescence for the SPAN-X protein demonstrated that SPAN-X was equally distributed between X- and Y-bearing spermatozoa, suggesting that SPAN-X mRNA and/or protein are shared within spermatid cohorts in the testis via cytoplasmic bridges [28, 29].
The mechanism by which SPAN-X is distributed to only 50% of ejaculated spermatozoa remains a fundamental question of great interest. The possibility that the presence or absence of SPAN-X protein might affect the fertilizing ability of ejaculated spermatozoa remains an intriguing prospect. One possible explanation for the observed disparity in SPAN-X content is that all spermatids in the testis contain SPAN-X and, due to incomplete sperm maturation, only half of the cells retain the protein by the time of ejaculation. Nuclear condensation and loss of residual cytoplasm occurs in spermatozoa as they mature in the epididymis [30, 31] and loss of SPAN-X in half of the cells may relate to this process. A second possibility is that SPAN-X could be expressed and transported within only half of the spermatid cohorts in the testis. This unique distribution pattern raises interesting questions regarding SPANX gene expression and function in spermiogenesis and tumorigenesis. Experiments are in progress to further determine the incidence, localization, and function of SPAN-X during spermiogenesis and transport through the male and female reproductive tracts and to follow the fate of SPAN-X during fertilization. Further investigation of SPAN-X expression may help elucidate the mechanism of SPANX gene activation during spermiogenesis and tumor progression, particularly in reference to malignant melanomas.
SPAN-Xa and SPAN-Xb Localize to Different Subcellular Compartments in Transfected Cells
Analysis of the CTp11 peptide sequence [5] showed 97% and 86% identity with the SPAN-Xa and SPAN-Xb peptide sequences, respectively, and absence of the six amino acid insertion found in SPAN-Xb. Transfection of mammalian cells with SPAN-Xa and SPAN-Xb mammalian expression vectors resulted in different subcellular localizations of the expressed proteins. Specifically, the SPAN-Xa protein was identified in the nucleus by immunofluorescence labeling, although the SPAN-Xb protein was largely found within the cytoplasm. The CTp11/eGFP fusion protein used previously in transfection of COS-1 cells was also identified in the nucleus [5]. The CTp11 protein sequence is most similar to that of SPAN-Xa, which also showed nuclear localization in transfected cells. These results indicate that the NLSs found in the SPAN-X sequence are effective in trafficking the SPAN-X protein to the nucleus. However, the SPAN-Xb protein apparently contains a variation that inhibits nuclear targeting in nongerm cells. SPAN-Xb and a homologous EST (GenBank number AA382424) were present in human testis cDNA libraries, demonstrating that the SPAN-Xb mRNA is transcribed in testis. Furthermore, the SPAN-X protein was immunolocalized to the nuclear envelope of spermatozoa and not to the cytoplasm. These findings suggest that the variation in the SPAN-Xb peptide inhibiting nuclear targeting in nongerm cells does not affect nuclear translocation in male germ cells.
On immunoblots of transfected cell extracts, the SPAN-Xb protein exhibits a higher molecular weight than that observed for the SPAN-Xa protein. Interestingly, the ORF of SPAN-Xb contains a consensus N-linked glycosylation site within the additional six amino acid stretch that is not present in the SPAN-Xa polypeptide. This asparagine may be glycosylated in the transfected CV1 cells, leading to the increased molecular weight of the 21–22 kDa SPAN-Xb protein observed by immunoblotting and to its localization in the cytoplasm. Studies are underway to test this hypothesis. The identification of the a and b forms of SPAN-X, their different subcellular localization patterns, and their different apparent masses in transfected somatic cells raise interesting questions regarding which form or forms of SPAN-X are expressed in malignant tumors and whether SPAN-Xa and SPAN-Xb represent gene products from a single polymorphic gene or multiple SPANX genes. Furthermore, these findings increase the importance of understanding the localization and function of SPAN-X in its normal cell type, i.e., spermatids and spermatozoa.
SPAN-X Properties and Cancer Immunotherapy
Expression of SPAN-X/CTp11 in tumors and normal testis establishes this protein as a new member of the family of CTAs [5]. CTAs have typically been identified using immunological approaches. The protein families MAGE, BAGE, and GAGE were defined by expression cloning as CTL targets of patients with cancer [32–34]. SSX, NY-ESO-1, SCP-1, and CT7 were defined by SEREX (serological analysis of recombinant cDNA expression libraries) that involves the construction of cDNA expression libraries from primary and metastatic human tumors, cancer cell lines, or testis and immunoscreening of these libraries with patient sera [1, 35–38]. CT7/MAGE-C1 and CT10 were defined by representational difference analysis (cDNA RDA) for genes with testis-restricted expression [39–41].
Unlike those CTAs identified using immunological methods, SPAN-X/CTp11 was found by comparing mRNA expression between a human parental melanoma cell line and its metastatic variant using mRNA differential display [5]. Thus, antibodies to SPAN-X have not been identified in patients with cancer nor have T cell epitopes been defined. The different subcellular localization of the two SPAN-X forms in nongerm cells raises questions regarding presentation of these antigens in cells that do not normally express this protein. For example, it is presently unknown if SPAN-X is presented in the context of a major histocompatability complex protein on the surface of tumor cells.
Although the function of other tumor differentiation antigens is known, little is known about the localization and function of CTAs in either normal testis or tumor cells. In the testis, members of the MAGE family of proteins are expressed in spermatogonia and primary spermatocytes but differ in their subcellular localizations [42–44]. SSX (HOM-MEL-40) protein, which is localized to the nucleus in transfected cells and spermatogonia [45, 46], contains a sequence involved in repression of transcription [47]. Translocation of this repression domain may contribute to neoplasia in synovial sarcoma. SCP-1 is the only CTA identified to date that does not map to chromosome X but rather has been mapped to chromosome 1p12-p13 [38]. SCP-1 is a synaptonemal complex protein that is expressed during meiotic prophase of spermatocytes and is believed to function in the alignment of homologous chromosomes during meiosis, the promotion of cross-over events, and chromosome segregation [38].
There is a new correlation that further emphasizes the importance of understanding the basic cell biology of SPAN-X. Previously, we mapped the SPANX gene(s) to Xq27.1. In the February issue of \i\Nature Genetics,\r\ Rapley et al. [48] mapped the locus of a susceptibility gene for familial testicular germ cell tumors, designated TGCT1, to Xq27. TGCT1 appears to be associated with a higher risk of bilateral TGCT and perhaps undescended testis (UDT) than other TGCT-susceptibility genes. Although other genes may be embedded within the TGCT1 locus, SPANX is one of the few genes identified in this region to date. Furthermore, the unique distribution of SPAN-X in 50% of human spermatozoa, its germ cell-specific expression, and its identification in human tumors recommends the possibility that the SPANX gene may be involved in the etiology of TGCT.
In relationship to tumorigenesis, the following features of SPAN-X biology may be relevant. First, SPAN-X/CTp11 shares several properties with known CTAs. For example, SPAN-X/CTp11 mRNA is expressed at high levels in several tumors including malignant melanoma and bladder carcinoma [5]. In the nondisease state, SPAN-X is expressed exclusively in the testis. The SPANX gene has been localized to the X chromosome, specifically Xq27.1 [6]. Genomic analysis of the SPAN-Xa, SPAN-Xb, and SPAN-Xc/CTp11 sequences indicates the possibility of a SPANX gene family. Similar to the CTAs MAGE-A11, SSX, and SCP-1, the SPAN-X protein localizes to the nucleus. Specifically, SPAN-X is a structural protein associated with the nuclear envelope of spermatozoa. Unlike the CTAs characterized to date, SPAN-X is postmeiotically expressed by haploid round and elongating spermatids during spermatogenesis. This is significant as postmeiotic mRNA transcription of X-linked genes is uncommon [6]. Furthermore, the SPAN-X protein is present in only 50% of ejaculated human spermatozoa, indicating a unique regulation of SPANX gene expression in the testis. In vitro, the SPAN-Xa and SPAN-Xb proteins target to different subcellular compartments, the nucleus and cytoplasm, respectively, possibly due to differential glycosylation in nongerm cells. Expression, differential glycosylation, and subcellular localization of the normally germ cell-specific SPANX gene product in somatic cells may have functional significance in tumor metastasis.
Several characteristics of SPAN-X/CTp11, including tissue-specificity and immunogenicity, implicate this protein as a good immunogen for cancer therapy. First, in 50 tissues examined, including 7 fetal tissues, SPAN-X mRNA expression was identified exclusively in the testis [6]. During puberty, new antigens, such as SPAN-X, are expressed in the testis at the onset of spermatogenesis and sperm antigens are normally protected from the immune system of both sexes. Thus, SPAN-X may be immunogenic in humans because tolerance is established in the neonate. Secondly, the immunogenicity of SPAN-X has been established in two mammalian species, mice and guinea pigs. All 9 animals immunized with recombinant SPAN-X demonstrated a strong immune response to the antigen with more than 19 isoforms demonstrating immunogenicity. Furthermore, insoluble antigens and sperm-specific nuclear antigens with properties similar to SPAN-X are often highly immunogenic [49–52]. The immunogenicity and microheterogeneity of SPAN-X demonstrated in the present study provides a foundation for the determination of immunogenic forms of SPAN-X in primates for use in cancer vaccine trials. Importantly, the availability of a specific immunoreagent to the SPAN-X protein permits the study of SPAN-X microheterogeneity, solubility, localization, and incidence in human tumors, while the recognition of SPAN-Xa and SPAN-Xb mRNAs opens the opportunity to similarly examine their individual expression.
ACKNOWLEDGMENTS
The authors thank Leigh Ann Bush for technical assistance, Dr. Mark Conaway for statistical assistance, and Drs. Pablo Visconti, Michael Coppola and Friederike Jayes for their critical review of the manuscript.
FOOTNOTES
First decision: 24 January 2000.
1 Supported by NIH grants HD U54 29099, P30 28934, T32 DK 07642, T32 HD 07382, U54 HD 28934, and D43 TW/HD 00654; and by the Fogarty International Center and the Andrew W. Mellon Foundation. 
2 Correspondence: John C. Herr, Department of Cell Biology, Health Sciences Center, Box 800732, University of Virginia, Charlottesville, VA 22908. FAX: 804 982 3912; jch7k{at}virginia.edu
Accepted: September 8, 2000.
Received: December 22, 1999.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  G. Almanzar, P. B. Olkhanud, M. Bodogai, C. Dell'Agnola, D. Baatar, S. M. Hewitt, C. Ghimenton, M. K. Tummala, A. T. Weeraratna, K. S. Hoek, et al. Sperm-Derived SPANX-B Is a Clinically Relevant Tumor Antigen That Is Expressed in Human Tumors and Readily Recognized by Human CD4+ and CD8+ T Cells Clin. Cancer Res., March 15, 2009; 15(6): 1954 - 1963. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Haraguchi, T. Mabuchi, S. Hirata, T. Shoda, T. Tokumoto, K. Hoshi, and S. Yokota Possible Function of Caudal Nuclear Pocket: Degradation of Nucleoproteins by Ubiquitin-Proteasome System in Rat Spermatids and Human Sperm J. Histochem. Cytochem., June 1, 2007; 55(6): 585 - 595. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V.A. Westbrook, P.D. Schoppee, G.R. Vanage, K.L. Klotz, A.B. Diekman, C.J. Flickinger, M.A. Coppola, and J.C. Herr Hominoid-specific SPANXA/D genes demonstrate differential expression in individuals and protein localization to a distinct nuclear envelope domain during spermatid morphogenesis Mol. Hum. Reprod., November 1, 2006; 12(11): 703 - 716. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kouprina, A. Pavlicek, V. N. Noskov, G. Solomon, J. Otstot, W. Isaacs, J. D. Carpten, J. M. Trent, J. Schleutker, J. C. Barrett, et al. Dynamic structure of the SPANX gene cluster mapped to the prostate cancer susceptibility locus HPCX at Xq27 Genome Res., November 1, 2005; 15(11): 1477 - 1486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. V. Harper and S. J. Publicover Reassessing the role of progesterone in fertilization--compartmentalized calcium signalling in human spermatozoa? Hum. Reprod., October 1, 2005; 20(10): 2675 - 2680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Harper, L. Wootton, F. Michelangeli, L. Lefievre, C. Barratt, and S. Publicover Secretory pathway Ca2+-ATPase (SPCA1) Ca2+ pumps, not SERCAs, regulate complex [Ca2+]i signals in human spermatozoa J. Cell Sci., April 15, 2005; 118(8): 1673 - 1685. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. V. Harper, C. L. R. Barratt, and S. J. Publicover Stimulation of Human Spermatozoa with Progesterone Gradients to Simulate Approach to the Oocyte: INDUCTION OF [Ca2+]i OSCILLATIONS AND CYCLICAL TRANSITIONS IN FLAGELLAR BEATING J. Biol. Chem., October 29, 2004; 279(44): 46315 - 46325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kouprina, M. Mullokandov, I. B. Rogozin, N. K. Collins, G. Solomon, J. Otstot, J. I. Risinger, E. V. Koonin, J. C. Barrett, and V. Larionov The SPANX gene family of cancer/testis-specific antigens: Rapid evolution and amplification in African great apes and hominids PNAS, March 2, 2004; 101(9): 3077 - 3082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. A. Westbrook, P. D. Schoppee, A. B. Diekman, K. L. Klotz, M. Allietta, K. T. Hogan, C. L. Slingluff, J. W. Patterson, H. F. Frierson, W. P. Irvin Jr., et al. Genomic Organization, Incidence, and Localization of the SPAN-X Family of Cancer-Testis Antigens in Melanoma Tumors and Cell Lines Clin. Cancer Res., January 1, 2004; 10(1): 101 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Wang, Y. Zhang, H. Liu, E. Salati, M. Chiriva-Internati, and S. H. Lim Gene expression and immunologic consequence of SPAN-Xb in myeloma and other hematologic malignancies Blood, February 1, 2003; 101(3): 955 - 960. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
